As an aside, there are tomes of literature, research and arguments for and against the computational theory of mind. I don't think it's as clear cut to say that the brain is a computer, as physiologically it certainly doesn't work like a computer at all.
On the other hand, I've met plenty of people who believe the brain has "processing power" that's equivalent to literal CPUs.
For example, the definition of "compute" can vary greatly, which we can see here in the comments section versus what is usually considered computing when it comes to the computational theory of mind. Often regarding this topic, "computing" means symbol manipulation and representation a la Turing machines, whereas many in this comment section interpret "computing" as the action of taking input and producing output.
My point with the comment is that I'm no expert, and I'm not someone who has really thought about or tried to tackle this problem. There are plenty of people who have, though, and they can make more insightful, apt and convincing arguments that are worth looking into before accepting or dismissing the concept entirely.
Saying the mind is somehow beyond quantifiable physics is equivalent to dualism.
So is there some animal brain, no matter how primitive, that we could in principle simulate? I think so. I think that is realistically within our capacity.
It also gets into metaphysics, as some people believe that mind is not limited to the brain.
However, given our current level of understanding of how the mind works, we can't say for sure that certain thoughts are algorithmic. For example, we don't know an algorithm by which new systems of axioms can be created, nor even a general algorithm for proving a theorem in a given system of axioms, beyond brute force searching. Given that human minds do these things relatively often, it's unlikely they are arriving at it by brute force search in the enormous domain of mathematical statements.
So, we can't say we've ruled out the possibility that the mind is using some unknown non-computational process for finding novel theorem proofs. Note that proof checking is algorithmic, only proof discovery is potentially not.
Now, to me extrapolating this to a non-computational theory of mind smells a lot like a God-of-the-gaps argument, so I don't buy it. But it is worth remembering the limits of what is actually proven to be computable.
I've taken this example somewhat directly from Turing's paper on computation - there, he bases the model of the Turing machine specifically on a mathematician doing rote mechanical calculation with a pen and paper (the machine state is their brain state, the tape is the notebook, the writing head is the pen, and the reading head is their eyes). However, non-rote calculations are arguably left out of the paper's claims (or at least most of them).
Much more famous arguments about non-computational theories of mind are, to me, much less convincing. The most famous is probably The Chinese Room, where the idea is that a man locked in a room receives a question on paper written in Chinese characters, and follows lookup tables and English instructions to return another paper written with Chinese characters, but can't really be said to understand Mandarin even if he is always returning the right answer for the question asked (or so the author claims). Another such famous argument is Mary's Room, where Mary is also locked in a room where nothing has the color red, but where she gets to rid everything that can be known about the physics and biology of the color red and its perception in the human brain (much more knowledge than we have today), but where she is assumed to still have a new experience once she leaves the room and actually sees red for the first time directly.
Notably, such examples wouldn't exist if computation was strictly a subset of thought, but that can be waved away by acknowledging that the brain has finite resources that it does its best to use sparingly. Also, brains definitely run algorithms with staggering numbers of variables, e.g. object recognition from noisy sense data.
And all the results about all computers being equivalent only work if they have access to infinite storage and power. That computation isn’t outside the set of all theoretical thoughts, it’s just outside the set of all ones in actually existing human brains that don’t have access to a pen and paper.
As to simulating a ball being thrown, that’s one thing that the brain already does. Visualization is a simulation.
Like if you're a bad shot you can't install an update to turn into a sniper. Of course, snipers do get to be that good by practicing enough, but I don't know if they visualize the target or do it with conscious math.
What’s important to realize is people don’t just picture the path a ball takes but also the feel as they make a throw. Predicting the path of a thrown ball to be able to catch it is a another sign of simulation, but combining that with a simulation of how a throw takes place is useful.
Pool is clear example of the brain doing physical simulation. The exact positions are unique, but people learn how to predict what will happen to figure out what’s reasonable to attempt.