Yes, I don't dispute the fact that there are some innate trait, more fitting brain structure, differences in sensory sensitivity (stupid example: perfect ear) that lead to better 'performance'. But a lot of said impossible tasks are named so because of a lack of understanding and thus a difficulty in teaching.

So far music school way of teaching is by sitting you on exercises while someone verify the output. That is not teaching per se, it's a black box system that has seen somehow regular results. Kinda like making glass, there's a procedure for that, but what really happen at the molecular level is still blurry.

With time the student's body and mind change and he gets skilled at it. How does he explain it ? does he have words to communicate his feeling to a newbie ? Never seen so so far. Recently some violin teacher was praised because he discovered non-linearity (the fact that inertia in normal speed gestures is not the same, thus learning technique requires you to fake inertia by allowing acceleration at slow speed, which means exaggerated movements). Still a shallow discovery to be honest. It doesn't really describe how to create and sustain this kind of inertia between multilimb chains and how to attempt it pleasurably.

Back to code, similar issue is present when one of the most effective (and pleasurable) way to learn the art is by pair programming (with a mentor or even a same level pair). It shows that some notions are still not taught properly in schools. Maybe the system is already at peak and there's nothing to change that would lead to better progress.