My answer won't be a Complete Guide (tm) or anything like that. These are just some rules that will help out.
Your professional journal should be ONLY stuff you would be prepared to let other people look at. Maybe stuff that is proprietary and requires a promise not to misuse it, for example. Like data from an experiment or your half-finished manuscript. But not personal or sensitive stuff like "human resources" type stuff. If you are recording stuff you would not want others to see that should be separately and more carefully recorded and secured.
These days you probably want your journal to be on-line. You probably want to include hyperlinks to web sites. At this office we have had some good experiences with some free "wiki" software that lets you edit things and view them and search them. There are a bunch of these. You can easily find review web pages that compare and contrast their features so you can pick out one for you.
Consider for any given bit of info whether you need to put it in your journal or cite a source. Maybe a cite with some explanation, a page/paragraph type location. maybe a couple quick notes to say why you need it. Depending, maybe a quick drawing.
Consider what you will remember in a year or two. You won't remember what you were thinking that led you to the spot you were. Or what the problems were that were holding you back. Or what the partial solutions were that you were considering. Think about the things that, in a year or two, you would like to still have available but that right now are only "in your head."
Think about the explanation you would have liked to have when you were first learning the subject.
Think about the explanation you would have liked when you started a new project in the general subject area and need to repeat some mundane sub-task.
Think about the explanation you need when you return to the subject. "How did I do that the last time I did it?"
These last few are probably very different. They don't necessarily go in the same spot in your journal.
Think about the "usual suspects." If you find you are repeatedly going to the source docs for a particular thing, you should have the standard source and maybe the standard info recorded in your journal.
Think about being able to get the info back out of your journal to use in your work. So whatever form it may take, you want it to be compatible with the usual way you create documents so you can cut-and-paste things like drawings, equations, and references to articles. And definitely if you put tables of data and such in your journal, you want to be able to conveniently get it back out.
And that means you need to preserve the source of stuff that needs to be cited when you use it. So if you got a figure from such-and-such, that cite has to be right there with the drawing so you avoid a game of "hunt the wocket" when you want that cite. Possibly you even need to record things like "can only be used if" type information.
And, now to get to the buried lead. If some thing caused you heart-ache to get through, with some hard won lesson at the end, stop and record the way to get through that. If you learn some abstruse method or some complicated math or logic proof that gets you through, record that. Stuff that held you up until you had the Aha! moment, record the thing that gave you that Aha! And enough of the process so that you will recognize the situation the next time it happens.
