This is such a good book it has taken me awhile to work through the book. All the while finding examples of why people should read the book.
The summary in 2.3 does a good job of explaining why this subject is so important. Easy to pick a model, hard to get it correct with reliable, trustworthy results.
I was asked what models were in the book. All the commonly used ones like K-Nearest Neighbors, plus models like Multivariate Adaptive Regression Spines and Cubist Regression Trees for Regression Models.
Classification Models including Nearest Shrunken Centroid and Nonlinear Classification Models.
Well thought out examples with the R packages and example code.
Take your time and work through this book.