The book is not hard to read. Each chapter is written independently and hence the chapters can also be read independently. Each chapter presents in an accessible manner some ideas taken from a couple of Lamport's seminal papers.

In this sense, I believe reading Paxos from this book (chapter "Chapter 4 State Machine Replication with Benign Failures") would be much easier than reading Lamport's "Paxos Made Simple" paper. Actually, chapter 4 is a great read even if you know Paxos.