I would agree that the two things that MUST occur when you process a project are:
1) Put it on your project list with a successful outcome, get all the clarity you can about what this will look like when it is 100% complete and succesful
2) Get a Next Action onto one of your context lists. (Often that action is "brainstorm/plan...")
Honestly, I have not done #2 when I was in a hurry. Just got it on my project list and caught it at my weekly review, but personally I have found that reviewing a project that has "stalled" in the system just takes the wind right out of my sails. Therefore I ALWAYS find a Next Action, if only to free up my mind from wondering if I am doing ANYTHING about XYZ. To know that the next step is on a list and that I have the freedom to choose to do it when appropriate is enough for me. For larger projects, I have actually put the planning/brainstorming time on my calendar so that I can get a jump on it. I've blocked out a few times a week that I can "park" action items as hard landscape so that I can add some urgency to particular tasks.
I found the hardet part was learning to trust that my system indeed had captured a particular item, particularly those projects that could cost me my job! That's another reason I always pick a Next Action, I have a reminder on a list that I will check several times before my weekly review. Gives me a jump of a few days.
Hope this helped!