One thing to remember about this book that is easy to miss is that it's not a memoir or autobiography in the usual sense. You may have seen this book, and its sequel,
What Do You Care What Other People Think?, described as being "by Richard Feynman and Ralph Leighton" or "edited by Ralph Leighton" or "as told to Ralph Leighton". It turns out that it was not written as a book, but his friend Ralph Leighton (son of the physicist Robert B. Leighton) took several hours of (recorded) conversations with Feynman talking to him, and selected parts of them to go into the book.
So although the book is sold as "by Richard Feynman" (which is true in some sense: it's in the first person, published when he was still alive, and he really did say everything that's in the book), it would be more accurate to call it a book by Ralph Leighton, and an appropriate title may be "Things my friend Richard Feynman told me about himself that I thought were fun". (This also explains the subtitle "Adventures of a Curious Character"—this is not Feynman calling himself that, but Leighton describing his friend that way.)
Now consider that (1) Ralph Leighton was not a physicist but Feynman's "close friend and drumming partner", and (2) Feynman was a natural conversationalist, automatically adapting his style depending on whom he was speaking to, whether it was a friend, or undergraduate physics students, or he was talking about computers to a New Age crowd at Esalen (https://www.youtube.com/watch?v=EKWGGDXe5MA), and you have this result. The friendly conversational style increases the appeal of the book and gives some insight into the speaker, but the technical detail of Feynman's core work, which was very important to him, gets diluted.
(One of my professors disliked the book for giving the impression that you get a Nobel Prize just living a life of having fun and playing around, while in fact Feynman was known to work very hard, at all hours of day and night. He liked to re-derive by himself anything he learned till he was satisfied he understood it, and that took prodigious amounts of pen-and-paper calculation. This may partly be Feynman projecting an aura of effortless brilliance, but I think it's more likely a combination of the fact that hard work doesn't seem hard if you enjoy it enough, and that going in detail about how hard you worked does not make for very good conversation.)
It is true that many memoirs are written this same way (dictating to someone else), but I think this book shows the effects more than most: including the selection of topics, as you observed.
(BTW the audio material that went into the books is available too, as "The Feynman Tapes", and listening to it may give a different impression than listening to an audiobook of someone else reading the text of a book itself transcribed from audio: https://kongar-olondar.bandcamp.com/ )