There are two stages to learning Git. The first is where you learn how to do basic CVS/SVN-like version control with Git. This you can get from a lot of the online tutorials and from the two published books you reference.
But if you stop there, you're missing so much. The second stage of learning Git is where you really get Git and see it as a toolkit that operates on a simple and well defined data model. The two published books don't do this stage much justice.
Be sure to checkout the excellent PeepCode PDF book (US$9) Git Internals. A free article along the same lines is Git From The Bottom Up. Both of these references will help you better grok the essence of Git ("content addressable data store", DAG, data model, etc.). That, in turn, will help open your eyes to the marvels of things like cherry-picking, rebasing, merging, etc. They'll strip away some of Git's magic and help you understand all of the cool things you can do with Git (many of which aren't in a simple, one-line command).
Update: The Pro Git book is now available in print and free online. That's an excellent book along the lines of the Git Internals book but complete with basic usage, git-svn
, submodules
, example workflows, etc. The source for the book is, of course, available on GitHub. Highly recommended.