Don't get me wrong the book is good, the discussion of numerous views, styles is great and it provokes a lot of structured thinking, it's a book I would recommend.
BUT, it could have been so much better... (hence only 4 stars)
1. Needs more example models & diagrams, a picture is worth 1000 words.
2. It should be about modelling not documenting, but hey that IS the title of the book!
3. The final section that discusses the views proposed by RUP, siemens, ODP etc are clearly an afterthought as they don't really tell you much and are so 'standalone'.
4. Amazingly there is no discussion of frameworks.
5. The MDA paradigm isn't even touched upon - a major ommission IMHO.
Ultimately though it's the best book I've read on the subject, just left me with half the answer...