A key distinction between OOP and Functional that I'm surprised is left out here is the way they handle state.

The author talks about functional languages being good for parallel processing and distributed systems, but a big part of that is immutability. When you don't have to worry about data transforming out from under you entire classes of problems become a lot easier.

Of course without mutation a different class of problems becomes a lot harder.