If I were you, I'd go to where that knowledge does get taught: in undergraduate mathematics classes. That usually starts with some rigorous calculus (I second the recommendation for Spivak), real analysis (which might seem hopelessly unmotivated if you don't know/remember much calculus), linear algebra or abstract algebra, or discrete math. Depends on what you're interested.
Reading mathematics is hard. Google around for advice. Do lots of problems, which generally means finding proofs. This would very well fit the definition of "discovering mathematics".