My pet theory of genius

My concise pet theory for what makes a genius goes something like this:

1. A degree of aptitude. Doesn't have to be Neumann-level. Beyond a certain threshold (+1 SD?) more things are in reach than most people assume. I believe you can get most people to pass Calc 3, given the time and resources. But this is basically what Feynman says, and do you really trust him to be an ordinary person?
2. A culmination of learning unimaginable to many people, a volume of work. If something is the top idea in your mind for years and decades, it's unavoidable you will become good at it. And it doesn't have to be years and decades if you put in the requisite intensity.
3. Hidden skills. These are skills, models, or modes of thinking that help solve problems, are not visible from the outside, and are often hard to teach (somewhat described by the term Tacit Knowledge). Besides volume and aptitude, I think this is what separates top-performers from most practitioners, who presumably put in a decent amount of volume and possess above-average aptitudes. I don't have good data on this, but some examples of hidden skills come to mind: Feynman's integration trick and Po-Shen Loh's description of math competition problems (where the problem-solving skill of Thinking in the Wild can become a useful, hidden skill elsewhere). Feynman's trick is a hidden skill that helped him solve many other problems, while problem-solving skill beyond the confines of 'textbook problems' is what separates the mathematical cream from the crop. Hidden skills explain the common saying that you can get to red (top 99.7%) on Codeforces knowing the same algorithms as a specialist—it’s all 'problem solving skill'. Related: What Univeral Human Experiences are you missing without realizing it?
4. Problem taste. Given that you can make a lot of progress, where should you direct it? Andy Jones became a leading AI researcher through his Scaling Scaling Laws with Board Games Paper, which addressed the leading questions of the field only a year after Kaplan scaling laws came out. Also see: Apprentice to Genius, Patrick Collison’s great conversation with Dwarkesh about the Arc Institute.
5. Contra #1, there are aliens. Edward Witten, the most respected physicist of our time, dominated mathematics while simply trying to invent the math needed to describe his string theory. He won a Fields Medal as a physicist. Maybe the above points can explain some of his genius, but maybe there are just aliens out there.

What is the takeaway from this? My best guess is: put in a volume of work unimaginable to most people; find and train your hidden skills. (You can sort of tell I'm more of a nurture person.)