Upon visiting this post when I needed some guidance, I am revisiting to maybe offer some baed on my insights.

After some trial and error, I found that a good combination that works for me is the following: I used Nakahara's "Geometry, Topology and Physics" because its thorough. It's a good read overall but not terribly exciting since it doesn't provide some extra intuition. For that, I supplied myself with Frankel's "the Geometry of Physics". Frankel contains a lot of illustrations and sometimes cover topics in a more intuitive manner. Finally, I found Baez's and Muniain's "Gauge Fields, Knots and Gravity" to be superbly written, so that it can be read very easily and gives a very strong intuitive (and physical) image of what's happening.

So, a combintation of those three was ideal for me.

As a side note, I briefly used Fecko's "Differential Geometry and Lie Groups for Physicists" but found its overeliance on exercises to be big negative. I love it when there are many exercises to go through, but the author sometimes relies on the reader to go through very important points through exercirses and sometimes the timing is off; when a new concept is introduced, it sometimes requires a change in the way of thinking about it, so to delegate the fundamental understanding of a new concept to an exercise that the reader might not be able to solve is not a very good idea. I use it from time to time to find alternative explanations and to solve exercises (many of which are excellent), so it's not all bad.

Another that I liked is Isham's "Modern differential geometry for physicists". It doesn't get into advanced topics such as those covered in the later chapters of Nakahara, but I found it very pedagogical and it excels at getting to the point fast and efficiently.