Whereas you ask some good specific questions about design, I would like to be a little more generic and add that the main thing for any hobby project is to get to something playable before your determination runs out or your hard drive crashes.
In a hobby/free-time-driven game development process, 99.99% of good ideas never lead to a game people can actually play. People in this thread have good specific advice - you should heed those you like. But I'd suggest not covering all bases immediately - planning is fine but you need to eventually do something, create something you can show off. A sort of save-point you can continue from should the unforeseen happen and you have to take a break from your project.
Keep your grand plans in mind but strip most of them away for your first version - you will often find even 'must-have' features are not so must-have once you realize other things are not yet in place. The 'perpetual beta' is a tried and true concept for a reason; not assuming (indeed knowing) that your first few versions will be incomplete and imperfect can be quite liberating and allow you to get something out sooner.