There are several big things you didn't mention, AI and neural networks, for example. You could also try robotics and engineering stuff (watching how a piece of silicon and metal becomes alive is priceless).
This is good advice, the problem is that you are suffering burn out. !0 years sounds about right. You need to get back to your roots and one way I recommend to do it, is to get into embedded. It gets you back into a small reward loop where little things net positive feedback. The other thing that I recommend to people is get closer to a research discipline. If you are really that burnt out, try applying your trade to a secondary science that you are interested in, chemistry, Bio-Engineering, astronomy something like that. There are a lot of cool projects that need help with code.
I've actually used Silicon Labs' MCU University kit, and I highly recommend it. I don't get any money directly from it, but Silicon Labs does pay my salary.
The nice part about their tutorial is they take you through all of the details such as building you own PCB and other items where some of the kits have all of the electronics built for you. It depends on your interest but I find designing the electronics to be as rewarding as actually programming a micro-controller. I am also the kind of guy that likes to understand ever small detail as well.
I start using PICAXE(http://www.rev-ed.co.uk/picaxe/) few years ago. It's incredibly cheap and easy to use. But i think it's essential to have electronic/mechanical knowledge in order to make interesting project.
Pick up an Arduino, or one of it's derivatives (like Netduino). Then span out. Once you've gotten 'in', you'll soon find other interesting platforms to develop on.
Strangely enough, if I had the money, I'd probably go back to university and study one of the splinter-subjects genetics must have exploded into during the past 15 years. I've kicked myself over the past years repeatedly for not staying in university and getting some research job there. Maybe I can have a look into the DIY aspect of biotech, that someone already mentioned.
there are countless others just use Google. This will help you establish a network of individuals in the industry. Software skills along will not get you that far. You need domain knowledge as well as contacts in the industry.
AI (neural networks/fuzzy logic/genetic algorithms) were actually part of my focus at university, and something I felt pretty exited about back then. That never went anywhere afterwards though. The realities in the job market here are pretty much anti-innovation for the most part ("We can't do that - we'd be one of the first companies in germany to work with XXX" is something I've heard pretty much verbatim more than once).
Maybe I should read up on my differential equations and give Kohonens "Self-Organizing Maps" another whirl, just for kicks.
