why have user context and kernel context ... unix
There is two main kind of scheduler in operating system, preemptive schedulers and non-preemptive schedulers.
Non preemptive schedulers would behave like you think, a process with higher rights and higher priority will keep using the cpu until it finish OR until it block (on a mutex for example or with a call to yield which explicitly release the cpu in order to schedule another one.)
But non-preemptive schedulers are rare and linux scheduler isn't that kind. It uses time slices to let process work for a short period of time before de-scheduling it, it also include priority but keep scheduling processes with lower priority, you should take a look at this linux scheduler article.
This Stackoverflow posting has a discussion that includes a run-down of how kernel mode works with an explanation of some of the jargon. In particular look at the section titled 'A brief primer on kernel vs. user mode'. It might help to shed some light on your question.