The introductory book by Ken Rothman (which will affectionately be known as "Baby Rothman" from here on out) is not a representation of the quality of Modern Epidemiology by Rothman, Greenland and Lash (ME3).
Baby Rothman is meant to be a very basic introductory book, of the kind suited to a class non-Epidemiologists are taking for distribution requirements, or as a first step to someone who hasn't encountered much Epidemiology before.
ME3 on the other hand is essentially the definitive reference book for most epidemiological methods. It is the only Epidemiology textbook I've had that has always come with me, regardless of the project I'm doing, and it's proved invaluable. There's more than a few questions I've answered here with citations from it.
Beyond ME3, a few of the books I use regularly:
Survival Analysis Using SAS: A Practical Guide by Paul Allison. If you're a SAS user (or possibly even if you aren't), its a very good treatment of the doing of survival analysis.
Survival Analysis by Klein and Moeschberger is a more theoretical treatment and reference on survival analysis, but makes for a good supplement to Allison's book.
Modeling Infectious Diseases in Humans and Animals by Keeling and Rohani, if you're interested in mathematical epidemiology, is a good introductory book that keeps a balance of practice and math.
Most other references I use are either very domain specific, or programming books.
But seriously, if you have to buy one book, that book should be Modern Epidemiology.
Best Answer
This book is really good, though very introductory: What is a p-value anyway? by Vickers. He draws on his experience as a medical researcher. Perhaps it provides a good high level of info and helps with the intuition side of things.
However, I know you asked for more of a reference book. I have heard good things about these 3 books from several different colleagues: