Using the Script component indeed solves the problem, but it's a work around not a solution.
Regardless I'll just throw in my suggestion about this and the reasons.
1. Commentary is the author's way of communicating with it's users.
When I use python, I write comments to save others from attempting to understand what's happening. It's very hard to read code without them and sometimes you just don't get what's happening.
In FilterForge, I also throw in some commentary just so that I can explain why I did it that way. Much easier to understand than not seeing anything. FilterForge has a decent reputation for being a programmable graphical interface, meaning it's not just a graphics package with buttons. We're talking "programming" here and not Photoshop.
2. It's a way to keep things organized.
In order to keep one's self organized I use comments. I use comments to show that's happening to myself. I title classes, definitions, and loops.
In FilterForge, the current way I keep my stuff organized is to simply drag out sub-groups of nodes. This is still, however, makes a mess. I would desire a much clear indication that I am dealing with a group / block of items that does something that I wanted it to do. I don't want to figure it out again.
3. It's a good way to have others understand the work.
FilterForge does NOT have a learning curve. Yes, we can plug-n-play. BUT more over, when new-comers arrive, there's nothing for them to see or learn. FilterForge as of this point has no "true" tutorial. I mean ok, there are a couple, but there's nothing that explains how things work, why they work and so on. There's NOTHING that explains why my node structure is necessary to the new comers.
So yeah. Basically the same reason as #1.
Regardless here's an interface of Quartz Composer. Note how detailed the notes are and how intuitive the interface is. We're not dealing with just nodes that make no sense. We're dealing with words not just nodes.