In contrast, Jeff's advice is the only way he found to work effectively and, he reasons, the only way anyone in a similar profession can work effectively. In reality, you may think that having next action lists and reviews are "productivity porn" and wasted time, while others may think that they help keep your mind clear and focused. You're going to do that work of evaluating priorities and deciding what to work on anyway, so there's no harm in being honest about it, isolating it, and minimizing the context switches. There's also no harm in going Jeff's route and doing "lazy evaluation" on that meta-work, only deciding when you need to. We all think and remember differently, so the debate will never truly end.
Personally, I'm in favor of advocating the best hacks and systems that people have found, preferably with grains of salt and low buzzword count, and letting people figure out what works best for them. As coders, we're used to the luxury of strong opinions and relatively hard evidence, but that's not something that will happen here.
Looking for a Korean tutor, completing that Javascript tutorial, or checking to see if there are welding lessons at the local community college are proper next actions that could be on a todo list.
I suggest looking into a quality system like Getting Things Done. You're on track with some suggestions (using pen and paper to dump tasks and clear up your mind, for example), but could use some refinement.