I have spent a lot of time this year battling the same thing. Tried this, tried that, tried just about everything under the sun.
And I realised something, only this past week: None of the systems I have looked at "suit", because I'm expecting the software to do the work for me.
I don't believe it matters what software package you use - project management still requires work on the part of the project manager.
I started planning our own, got to the point of doing up some wireframes and workflow charts (and got really excited about it), and then the obvious dawned on me. I will still need to spend time managing my projects, even if we built our new whizz-bang system.
We're using a heavily customised version of NetOffice, and it certainly has it's flaws, but ultimately the reason it's not working comes back to me. I'm expecting everything to just happen, and that will never be the case.
A large part of project management is communication, and I have found that trying to get software to do everything you want actually hinders that. So, if I need to know what the guys are doing right now, I ask them. We sit down at the start of the week and look at what's on, discuss priorities, issues, and so on.
I have decided not to waste any more time on trying to find or build the perfect project management software. Having a good system is more important than the software.
Perhaps this doesn't apply to you Mitcho - but before you waste as much time as I have, think about how much do you really need the software to do, and you much can you achieve with some coffee and a whiteboard.