Here’s a diagram that might help.
Dashes indicate optional data or processes. The “fit” derivatives include a collection of individual volumes and transform files. The “transform” section shows the process used to generate resampled BOLD series. The available inputs, such as fieldmaps and slice-timing metadata, and the target space, such as an MNI template, determine the final result.
One important point is that we try to resample as few times as possible. So while we do always resample to native (STC + HMC + SDC), that is used for processes that need data in native BOLD space. When we resample to T1w space, we apply (HMC + SDC + BOLD-to-T1w) in a single step from the STC’d (if applied) time series.