Does RL give computers an “imagination”?

Recently on Hacker News, there was a discussion on Nikola Tesla his method of coming up with inventions by visualizing machines and their individual moving parts. [1] Do you think this is comparable to how AI "simulates" possibilities through iteration? Also could we learn from people like Tesla in improving RL? [1] https://news.ycombinator.com/item?id=35992553