Ah, Tristan…if I may ask: what are you smoking, and can I get some before flu season? 
A zombie close enough to a nuclear blast to be set on fire isn’t going to be a “walking molotov” threat to anyone. For the simple reason that it’s any humans (or buildings) close enough to encounter one will have been killed by the blast themselves.
A lack of a fear of fire or pain might actually be beneficial—with a big, pretty bonfire making an ungodly roar, and a 40 mph gale at his back, a ghoul might actually walk into a firestorm. And as the thread I’ve cited earlier noted, in one medical examiner’s opinion, a body can be burned to physical immobility in even an ordinary house fire in about half an hour.
As for the pressure wave…well, a 1.2 megaton blast (that’s the highest listed yield setting of an American B-83 strategic bomb), everything within an almost two mile radius is going to be hit by an 20 psi overpressure blast wave. That’s about 500 mph, by my numbers. I’ve seen figures claiming that somewhere between 10 and 20, limbs start getting blown off. At five miles, it’s about 5 psi—about 160 mph. Now, according to the handy old nuclear bomb effects computer, that’d be enough to fling a man in an open field at 25 ft/second, enough to cause a skull fracture—which Brooks himself lists as the minimum required force to re-kill a zombie brain—on impact with a hard surface. (I’m leaving out debris injury—either to the brain, or simply dismemberment by flying debris.)
At this distance, a target would also absorb a bit under 50 calories per square cm of thermal energy; according to one source, 19 cal/sq cm is enough to burn dry pine. Denim ignites at 27 cal/sq cm. At 10 miles, a target would only absorb a bit under 10 cal/sq cm. Not quite enough to ignite clothing, but enough to ignite newspaper, and easily leave a human with a 2nd degree burn. The possible effects on undead are somewhat less certain—I can’t find exact details on what kind of energies are needed for flash blinding, or permanent eye injury. And on the whole the process would probably be more analogous to cooking, or drying. Perhaps a chef would be able to weigh in!
For the record, according to my numbers, a nuclear radiation exposure 5000 rems (50 sieverts) is required to cause fatal neurological effects. There would only be a 500 rem/5 sievert exposure at 2 miles from a 1.2 MT airburst.
On top of everything, we have to consider the circumstances of the weapon’s deployment. If your strategy is to lure the zombies into the center of a city with noisemakers, THEN nuke them, they’re going to be heavily concentrated, much more than a populated human target. Probably more so than normal living people would would ever group, even in a large mob, riot, or outdoor concert. That’s going to reduce the number of merely “wounded” zombies even further. I mean, look at this blast map of New York. Or San Francisco. How many ghouls are going to cram themselves into the primary kill areas alone, with a big enough lure? They’re not going to wander off, flee to the countryside, or care about trampling other zombies (or being trampled) to try and get to the pretty noise.
In other words…stir-fried zack. 