Right after high school, I enrolled in a d.e. course at the local junior college (ok, I was a masochist). We used the Zill text, although not the boundary value problem edition. Needless to say, that book was a godsend b/c the instructor was horrible, so after awhile, I only showed up for class for exams, and self-studied on my own from that book. I recall that the book was fun and easy to understand.
Why is it good? It explains things in clear language. The proofs are laid out clearly. There are lots of example problems with solutions. This was critical in the portion of the book where he explains how to solve d.e.'s with variable coefficients. The book makes differential equations look interesting, which is important to capture readers. Zill also has a calc book, and mygoodness, that book was sort of repulsive b/c of the 70's style printing and the nasty brown colors. Looks are always a big thing, back then and now.
I'm not sure how applicable this text is for hard-core math majors, but definitely, if you are in engineering and don't require any weird esoteric understanding of the proofs that math people might need, this is text worth referring to. I can't comment on the BVP, though... However, it helped me to earn an A+ at Cal that first semester as a freshman, so he must be doing something right.