I don't know what a "genius" really is, unless it just means someone who's a) prolific in an intellectual field, b) releases high quality work and c) maybe contributed to a fairly "revolutionary" understanding of their field.
But it's unlikely that Feynman is "the last great American genius". Maybe the most well-known example is Chomsky (but there are no doubt many others), and he answers his emails quickly. So, it's not like we necessarily have to pore through notebooks and quips of dead people, guessing at their mental states.
Except that generally, the average person knows little about the inner workings of things.
It is easy for the average person to believe that building Twitter was hard, though surely it was. Our ability to understand things decidedly maps to things we can already understand.
To give a completely made up scenario about how this might have gone:
Stupid person: "Wow. Building twitter is hard. You type on a website and something comes to my phone. AMAZING."
Average person: "Eh. They probably just use some sort of service. I mean, if I can send a text, why is it amazing that somebody else can?"
Smarter-than-average person: "Twitter is down again? WTH? How hard is it to keep a website up? My Wordpress blog has never crashed, and all the posts in it are WAY more than 140 characters."
Smart person: "Yeah, I can see how scaling to x-thousand reads and x-hundred writes per second IS a big task. I guess you have to put in a really beefy database server to handle the writes and distribute the reads out to slaves, then do all your queries from there."
Smarter person: "If we get rid of relational database stores altogether, we can scale these records much better, and our only upper limit is memory."
etc.
Yes, it's a completely contrived example and I'm sure I probably insulted everyone who's ever done any Twitter-based naval-gazing, but that's what it is.
In short though, the average person is more easily fooled into thinking things are brilliant when in reality, they're made by equally average people who have studied in that particular field more than they have.
I think you are on to something here. I have worked with many very smart men and women who couldn't produce. I have know doubt that they were smarter than me (more knowledgable, quicker to answers, generally quicker to understand techniques and maths). Despite this, they would spend absurd amounts of time on fairly simple stuff. There was some need to "do it the right way" that led to effective paralysis. I'm not trying to diminish the importance of doing "it" right, but sometimes it is easier to just iterate than get the best solution from first priciples, particularly when we don't know the "right" way to begin with!
Upon talking to those folks, while the never stated it, I came to the conclusion that they were afraid of being wrong, or failing. I do my best to lead by example and fail spectacularly on tasks, only to learn and come up with a good solution when I can, but I am not sure how to communicate that failure on an iteration of 3 can be a huge positive. Even communicating that fact explicitly doesn't seem to mitigate that fear.
I suspect it comes from schooling, where incorrect answers are punished (with bad grades, peer judgement, teacher judgement, parent comments, etc).
I am working with some of those types of people right now. Researchers, who are undeniably smart, but seem to be far behind on their project goals, micro-managing incidental tasks, and not focusing on the big picture.
As to the source of the fear coming from schooling, I'm not sure. I'd never punish one of my kids who got a bad grade because he applied some creative approach or reasoning that happened to be incorrect, but most cases of bad grades I've seen are due to simple lack of effort.
But it's unlikely that Feynman is "the last great American genius". Maybe the most well-known example is Chomsky (but there are no doubt many others), and he answers his emails quickly. So, it's not like we necessarily have to pore through notebooks and quips of dead people, guessing at their mental states.