First, my post specifically said “reciprocating aircraft engines” – which were the only powerplants available when the typical airplane shape came about.
Second, imagine a jet engine (not a turboprop) in the nose of an airplane. It could get a little uncomfortable for the pilots and passengers.
A jet is a ‘pusher’ only because there is no way to make one a ‘tractor’. The engines either have to be out on the wings, or they have to be in the fuselage with the exhaust going out the back. There’s simply no way around that.
But jets don’t suffer from the same problems. For one thing, you can still taper the fuselage, and you don’t need a long driveshaft to an aft propeller. So you can put the thust line right where it needs to be. The intake for the jet is situated up front or up high, so FOD being sucked into the intake isn’t as big a problem.
In short, almost all of the problems that pusher airplanes go away with jets.
Just a semi-hijack, but was Louis Blériot the first to use the now conventional arragment of a monoplane with tractor engine, pilot-sitting-upright, rear yaw and pitch control surface arrangement?
Cooling is one of the biggest strikes against a pusher. You don’t have to worry about cooling airflow for a jet.
Cooling doesn’t seem to be a problem with any pusher design I’ve seen. Aircraft in flight need remarkably little intake area for cooling. Most pusher designs need no more than a low drag NACA style intake for cooling air. You’ve probably seen them on race cars, they look like a mostly triangular shaped hole with a point forward.
Cooling is indeed a problem. Some pushers have used belly scoops like a P-51 Mustang, others have used scoops on top of the fuselage, and some use water cooled engines with radiators somewhere up front. But it is a tough design problem.
One thing that I haven’t seen mentioned here and which I always assumed was a reason for using a tractor rather than a pusher: lift is generated by airflow over the wing surface. If you have a wing-mounted pusher, then at takeoff the air is not accelerated until it is behind the wing. On a wing-mounted tractor, the air is accelerated in front of the wing before it passes over the airfoil, which would generate additional lift.
I should point out, though, that I’m not an engineer, and I’ve never seen this put forth as a reason for preferring a tractor engine. It is just an assumption that I made.
The Rutans have made a number of modern pushers.
Having flown a Long EZ, I can tell ya that there is absolutely nothing wrong with the way a pusher flies.
In fact, I would put that down as the most fun plane I have ever moved a stick in.
But once upon a time, there were no NACA intakes. (Also, no NACA.) When airplane design was pretty-much conventionalized, the engines (which were often radials) needed to be in front for cooling.
NB: The P-51 was liquid-cooled.
As for the OP, we can use modern materials and modern engines to make very efficient, good-performing pushers; but “back in the day”, tractors were the most efficient configuration.
Sorry, I’m unfamiliar with the pusher version of the P-51, a rare variant of the type I’m sure. 
Please note the P-51 had a massive V-12 engine in a very small airframe giving it a very different power/size ratio than general aviation aircraft. Consider a plane about the same wingspan and legth of a Cessna 150 but with fifteen times as much power. That Merlin is going to need a smigen more cooling air than that little Continental. The P-51 also had a liquid cooled engine which is isn’t too common these days. Far from being a problem the scoop and radiator arrangement with its varible outlet actually added thrust.
I think that’s really the meat of it. Radials were the doiminant engine type until well after WWII. They had a good power to weight ratio but lots of frontal area. The NACA ring cowl came out in 1929 and managed to both reduce drag and improve cooling and for a couple of decades more that was sufficient.
Cooling is a problem with pushers. Usually, a very big one.
Aircraft engines need a lot of cooling air, to the point where cooling drag is a significant part of the overall parasitic drag of the aircraft. Most pushers end up using large scoops for cooling air. I assume you have seen the front of a Cessna 172 up close. How big (or how many) NACA vents would it take to flow the same as those two huge holes in the cowl?
I am a huge fan of pusher aircraft, and I read as many pusher articles/flight tests/builders notes as I can. Here is a good example of a builder trying to cool his example of the most popular pusher model made:
I’m sure he meant: “Some pushers have used belly scoops, like the ones on P-51 Mustangs.” 
FWIW, I like the Long-EZ. I went to Rutan’s shop and watched the presentation, but I didn’t have the money or space to buy and build the kit. 
They didn’t offer it as a kit, did they? I thought they were all plans-built. They sure are neat airplanes, though.
I want one of these:
Isn’t that what the P stands for? The Pusher-51?
They used to be offered as kits, but they stopped many years ago. AFAIK, they you can only build them from plans now. Burt Rutan is heading up Scaled Composites, so he has more lucrative things to do than the homebuilts market.
Yeah, I totally dig the Berkut. I saw it at a couple of airshows, and it’s very pretty. I think the fully-retracting gear make it look a lot sexier than the EZs.
Rutan’s designs were offered as plans - Aircraft Spruce may have offered “kits” in the sense of “here’s everthing you’ll need to build the thing”, but Rutan never sold any substantial pre-formed parts.
Rutan stopped selling plans because he got tired of being sued (thanks again, lawyers!), but, last I heard, RAF was still supporting the designs.
p.s. - it is illegal to use a “used” set of plans to build a plane the plans are a license for one plane.
The Cozy (which is a wider version of the Long-ez) is, I think, still available.
I am friends with a guy in the process of building the Cozy as we speak. Went flying with him in his Comanche last week.
Guys, I wasn’t saying that a P-51 was a pusher. What I should have said was that some pushers use a ‘P51-style’ belly scoop for cooling, like the Berkut. Others use scoops mounted higher up on the fuselage, like the Cirrus VK-30.
The Rutan Long-EZE is a great design, and one in which the pusher configuration makes a lot of sense. Because it’s a canard, there is no reason to either mount the engine above the fuselage (creating thrust line problems), or in the middle with a long driveshaft (creating vibration, weight and balance, and prop clearance problems).
Also because it’s a canard, you don’t have the control irregularities from having the prop blast over the tail in some attitudes but not others.
But it may still be that the pusher configuration is less efficient than a tractor would be. It’s just that you can’t mount a tractor prop on a canard. So once you go for the canard configuration, you have to use a pusher engine.
Realize that we’re talking about fairly trivial differences, which can be swamped by other design configurations. For example, most small seaplanes have pusher engines mounted high on pylons for the simple reason that they need to keep water out of the prop. That one design goal outweighs all others when determining where to put the engine.
I’d love to have a Long-EZE for cruising, but they do have their limitations. The biggest problem with a canard is that they need more runway than a similarly-powered conventional airplane, because the canard is designed to stall before the main wing. That means that to keep the canard from stalling on landing or takeoff, the main wing has to be at a lower angle of attack than pure efficiency would demand. Having owned an aircraft that was a bit of a runway hog, I find that a big limitation.
The most fun airplane I’ve ever flown was a [Glasair III]Glasair III](http://www.phoenixcomposites.com/N300ML.htm). 300 MPH, 3500 fpm climb rate, and it rolls on a dime. Great fun.
But here’s my idea of the nearly perfect homebuilt aircraft design: The Van’s RV series. They are cheap, conventionally built, metal, easy to construct, and really efficient. How about a 200 mph cruise, a stall speed of 49 mph, a takeoff distance of 270 ft (450 at full gross), a landing distance of 300ft (500ft at full gross), a climb rate of 2000 fpm, and a range of over 800 statute miles.
I don’t think there is another aircraft made that can match this overall combination of performance. There are faster planes, but they don’t land as slowly or use as little runway. There are cheaper planes, but they are not nearly as fast. There are plenty of more exotic planes, but this one you could actually have a standard A&P work on if it broke down in a strange airport. An RV will hold its value better, because it’s fully inspectable after its built, unlike many composites where you have to hope the builder knew what he was doing because you can no longer look inside the structure.
And, they are cheap. In an era where even a lousy Cessna 172 is pushing 200K, and the higher-performance homebuilts cost easily $100,000 to get into the air, you can buy an RV-6 kit for $13,000, and get one into the air with a used engine and VFR instruments for $25,000-$30,000.
There’s a good reason why the RV’s are the most popular kits of all time. And they look totally conventional, proving that you don’t need fancy canards and pusher props to get high performance out of an airplane.
Oh, and the RVs aren’t just the most popular kitplanes of all time, they are now one of the most popular airplanes of all time, period. As of the end of last year, there were 3100 RV’s completed and flying! Most other kitplanes can’t count even 10% of that number. There’s probably twice as many under construction. That puts the RV up in the range of the most popular factory-built aircraft, and the rate of sales and construction is higher than any other personal aircraft today.
Good design is often the simplest.
