Yeah, that makes sense. 
I over-complicated it by just quickly refactoring something I was working on.
Oh, BTW here is a screenshot example on how to call and make functions with the new nodes:
The Call Function node takes any kind of object, probably a Haxe or node trait. And lets you call the given function with arguments.
The Function and Function Output nodes let you create functions in node trees that can be called from other nodes ( using the Call Function node) and even from Haxe traits.