If the project is critically important, then you'll probably end up working on it all the time, so scheduling would only serve to help you make sure you're doing the most important things first. In this mode it's all hacking, you're just putting tasks in order.
Otherwise, scheduling is a very good idea for different reasons. If it's a personal project, which means there's no one else to goad you into working on it, then it's especially important to use scheduling to help you keep working. Remember what PG says about personal projects: Always produce. Do something everyday.
IMO a fixed schedule is always a better idea, and any hours beyond that are a bonus but in the end it all boils down to a matter of choice or habit. you should follow whatever pattern you are comfortable with
moonlighting makes it easy and hard at the same time. I can only work when I'm not working at my day job. I must work when I'm not working on my day job.
I started doing that "don't break the chain" thing. I would say I can spend around 18 hours a week on the project.
I am in a similar situation. Moonlighting too, but that should end shortly as I leave my current company for working on the startup full-time. Surprisingly I am doing the 'don't break the chain' thing too. I seem to be getting about 20-30 hours on my project.
Otherwise, scheduling is a very good idea for different reasons. If it's a personal project, which means there's no one else to goad you into working on it, then it's especially important to use scheduling to help you keep working. Remember what PG says about personal projects: Always produce. Do something everyday.