Here is a small remark about the Feynman lectures which might help people who are starting on them. When I first started using them I was informed of their legendary status and so tried to use them as a text for a long time.
Unfortunately I think this is something which is impossible to do-The Feynman lectures cannot be used as a text. There are a number of reasons for this.
Firstly a good text should have a certain number of problems and questions. While a supplement to the lectures is available these problems are not really sufficient.
Then the presentation is not systematic. If you are of a slightly mathematical bent of mind where you expect things to be developed systematically starting from a few basic axioms, then the presentation is extremely confusing.
Finally most results are derived in extremely unconventional ways. For example Feynman uses Carnot's argument on reversibility to derive a formula for gravitational potential energy.
All this is not meant to say that the Feynman lectures are bad. They are invaluable and the third volume in particular gives fantastic insight into
Quantum Mechanics. Also Feynman will tell you things which no other book will. However I believe that they must
not be used for an introduction to any topic. Once you have been exposed to a topic and have understood it to some extent then it would make sense to read what Feynman has to say on it which would then provide you with a new interesting perspective.