But this is the "100 paper cuts" thing. I use and curse Gnome daily. I could list ten more if I had any belief there'd be action. $1000 is a week's worth of action if you sell yourself really cheap.
Rant 1: If you look at a given "bug list" for any large project you'll find a dozen problems posted where the reply is a very curt "will-not-fix/not-a-bug/your-configuration-is-wrong". It feels generally snarky and unwelcoming, especially when I'm searching for a solution to the same bug. If someone could figure out protocol that would make bug-reporters feel more welcome, it would really help the process. As developer, I know that, in fact, a lot of things people see really aren't bugs but expected behavior/configuration problems. BUT!, BUT... you have to create an atmosphere of involvement if you want people involved.
Proposal/Rant 2: The format for menus in Gnome is the convoluted piece of trash that anyone ever foolish - how freedesktop could possibly publish it as a "standard" is beyond me. It needs to be junked and replaced with a simple approach. (And suspect this format is why panels has notable, annoying delay to be displayed even on modern hardware that should show stuff instantly).
Proposal/Rant 2: Scribus has a zillion bugs I'd love to see fixed. I have a private list I keep out of frustration. I won't bother pasting it here unless I see more interest. It's more likely that Inkscape will become a viable DTP project through supporting multiple pages than that Scribus will ever stop being a piece of total garbage - because it's really hard fix large, poorly coded project. Yes, I've used Scribus, a lot sadly.
The "recently used documents" seems like the most specific bug / fix that we might be able to get done first.
However, am I right in thinking that what you are seeking is essentially this:
If so, it looks like it has been fixed (or am I misreading the bug)?