I conjecture that, given any model, M, which conceptually goes to infinity, when M is applied to the real world, and its results are correlated to empirical observations, as M advances in the direction of infinity, eventually the observed nature of the real world results changes, and M no longer even approximately applies.
“Infinity” should be taken to include any infinite or infinitesimal meaning, including infinitely large, infinitely small, infinitely many, and infinitely long or short in duration.
I think this conjecture is saying two things: that any model has its own, finite, domain of applicability; and that nothing corresponding to our idea of infinity actually exists.
‘Course, I could be wrong.