The two objectives of maintaining a schedule is to 1) know when to bring resources--both human, materials, and money--to bear and 2) build a capability to know how you are progressing in both actuals compared to plan and your prediction of where you are headed.
The questions you are asking are the right things to ask but it really comes down to WHAT MAKES SENSE TO YOU so that you can maintain control. I have seen so many approaches that work so your approach really needs to evolve through experience, where over time you develop the one that fits your mental model such that you can safely say your project is under control.
My approach is to keep it as simple as possible, where I overlay the WBS into the schedule. My leaf level WBS is what I consider the work package and I may or may not decompose the package into lower level activities, depending if I think it makes sense. Some types of activities I would use a separate calendar, such as JAD sessions, because they can shift a day or two at a moment's notice, you can add or delete as you need them, and putting them into the schedule that is supposed to be under change control simply does not make sense. It would quickly not represent reality since reality can change tomorrow. And sessions moving around, adding or deleting, does not affect the baseline at the work package level.
But that is what makes sense to me. It is extremely important you develop an approach that makes sense to you. Trial and error is the way to go.