And also to the extent that mathematics matches our physical universe, there will always be experimental results which are undeterminable from known previous results, and which can only be learned by performing a new experiment.
I am sure scientists typically avoid assuming the human mind is up to the task of comprehending true nature and reality, which of course implies in totality (as a partial picture is not comprehension of truth). Those of us who have thought about this very question (which I bet is nearly all) would have to more or less realize there’s no reason to expect that the nature of the universe and reality would have a complexity limit, much less a limit near at hand; nor that the human mind, which does come at some evolutionary cost, would have been equipped with the surplus power to do this.
I sit here with my beloved cat purring behind me on the chair, our customary online pose. I am sure he will never comprehend integration by parts, or how the mysterious inner interpreter of FORTH can be a single microprocessor RTN instruction, or why helium might exhibit second sound. I also doubt he can see that he won’t. And I’m sympathetic, because on the grand scale we’re not very different.
Which experiments we perform, where and how and why we decide to look, and how we interpret the results, all depend on our theories.
And the theories we come up with depend on our world-view and the capacities of our minds and imaginations.
Yes, and I can guarantee that 99% of all humans will never understand those things either. I’m pretty smart, and I kind of remember doing integration back in college, so I suppose I might be able to learn it again if I tried hard, but I don’t feel like trying hard, so that’s out. At least I can help my kids with their algebra and geometry homework without looking like an idiot.
And even among the humans capable of understanding such things, none will ever understand all such things, because there’s just too much to know. And yet there are functional systems that depend on more knowledge than any one person can know, in order to operate.
Fewer. Not less. Fewer.
Point of clarification,
If you pay attention close enough you can generally guess the education level of a writer on the subject of physics by how often they use qualifying remarks.
The usage of words like [could, should, may] generally increases with education or knowledge.
Physics is descriptive not prescriptive. These observations do not happen because of these equations, equations describe our observations.
While there may be reasons for making absolute claims when teaching in k-12 or in other situations, you can generalize the idea that an expert is an expert because they realize how much they do not understand about a subject.
While science is commonly used as a noun is is really a process and should be a verb. Focusing on what the process itself produces and not the process misses a large portion of what makes science one of the most valuable human inventions. It is merely curious people trying to find out more about our world.
Even with less hard pursuits I am unaware of any topic that has been chased to a point of full understanding. I would argue that if anyone think that they fully understand any subject that is merely a mark that they have reached the limitation of their creativity.
It is not that no more questions remain, just that they are unable to come up with new questions themselves.
That’s a pet peeve of mine, too.
While I can try and empathize with prescriptive grammarians and this pet peeve as this is GQ I want to point out that less has been used for countables in English for over 1000 years and even by Alfred the Great.
Sure Robert Baker who is obviously one of the premier personalities of Prescriptivism preferred it but it does not seem he is demanding it.
While I realize popularity of this preference has risen recently, even countables may be considered as quantities rather than numbers in some contexts. I would argue it that is possibly the case here.
Perhaps this is because the unidiomatic hypercorrection of “one fewer” is a pet peeve of mine.
To clarify,
In both historical and present-day usage; fewer is restricted to count nouns but less is often used with count nouns.
Baker’s preferences simply conflicts with the typical usage by native speakers. Hopefully that information gives you one less problem to worry about. 
So, is < the “fewer than” sign?
Few people would say so.
I’m wondering if over time questions follow a bell curve shape almost where the X axis is time and the Y axis is unanswered questions.
At first the more questions you find and answer, the more that come up. But the universe (and everything in it) is finite in complexity, so over time the number of unanswered questions starts to decline. However who knows when the hump occurs.
Look at it this way–if iconic SF actor Matt Frewer were in an accident that resulted in the amputation of a limb, there would be less Frewer. If he were to die from his injuries, the world would have fewer Frewers.
There is another aspect of this, cosmologically speaking, in that as the universe expands in an accelerating fashion, there is more of it that we will not be able to access (at least, without superluminal travel or non-local connectivity), so as time goes on we’ll actually know less and less of the physical world.
As Chronos notes, our brains–which are evolved for the tasks of recognizing facial expressions, finding food, and making simple tools–are already overtasked with understanding the modern world and the complexities of it notwithstanding the fundamental mechanisms of physics which defy our everyday experience of causality to the point that people often describe aspects of quantum mechanics as paradoxical and we can only describe it accurately using mathematics of probability. Everything from systems biology to modern finance requires the use of computer modeling systems to make all but the most trivial predictions of cause and effect, and we are already at the point where sophisticated heuristic systems (complex neural networks) are doing things in ways researchers can’t explain.
The use of computers for visualization (i.e. representing data in conceptual or abstract ways that lets our monkey brains comprehend it in a limited fashion), prediction, and control allows us to expand our abilities, but it is wrong to think that it gives us a true comprehension of the information. As an analogy, you can draw a picture of a bicycle, create a free body diagram, describe the mathematics of how it operates, but none of that makes you able to ride a bike without the oft-painful learning process of falling off of a bike until you learn the complex balance. Comprehension of the more complicated systems of the world is like riding hundreds of bicycles at once; we just are not physically equipped to do it, so at best we can ride one bike and invite friends to ride the others together.
In fact, the math we used to describe the physical universe is probably just a poor approximation of what is really going on, and it may be beyond our ability to really accurately describe the underlying mechanisms and make predictions without actually creating a simulation which is as complex as the universe itself, which is thermodynamically impossible.
In terms of the actual questions we have about “the universe”, it has grown by such an immense amount even in the last couple of decades that we now have not only more data than we know what to do with (so-called “Big Data”), but actually too much data to practically categorize by human effort alone (“Colossal Data”). We are going to have to build artificial cognition systems just to be able to even identify the right questions to ask, and that will inevitably lead to even more questions. Those people who complain that we are at an “end to physics” just because we haven’t discovered new particles beyond the Higgs boson (which is not strictly true, even if we cannot access the energies for supersymmetry) are ignoring that we have so much information now we can barely even formulate the right questions to ask, much less run all of the experiments and observations to find answers to them.
Stranger
Is Frewer’s compressibility limit determined by maximum headroom or minimum legroom?
Stranger
For the Friedmann–Lemaître–Robertson–Walker or the de Sitter–Schwarzschild metric? 
With the understanding that there is no personal judgement being made this is a good example of my previous statement.
Us humans tend to disfavor ambiguity and we typically error to concrete rules. I am sure that there is some reason that this was probably selected for but I am unaware of what that would be.
Just as in Physics where undergrad and classical physics often speaks of laws and truths language enthusiasts and writers sometimes tend to prefer prescriptive grammar. People who tend to focus on the study of language are rarely prescriptive and tend to focus on the descriptive aspects.
One of the beautiful aspects of the scientific method is that it provides a framework to move past these basic human tendencies.
And getting back to the OP.
Even if you are an experienced astronomer and you search through a bunch of astronomical images like from Hubble you’re going to occasionally come across a “What the bleep is that?” stuff. A lot of weird things are discovered each year.
OTOH, the number of weird things that are explained per year is quite low. Sometimes you get lucky and a whole group of odd stuff is seen to be caused by a certain phenomena, but there are many one-offs. If you are an astronomer (and more importantly if you are grad student), it’s the group explanation results that matter the most. That will get you fame and fortune. Well, maybe a beer at a conference. So the one-off things tend to pile up quite quickly with very few being explained.
A lot of fields are like that.