The secret is not only formulating rules where things can go but also where they can't go. This leads to further deductions and a more complete diagram.
I am going to do a tutorial on this soon, hopefully a youtube video once I can get my camera situation figured out.
A last resort strategy is to do something. Getting flustered and worrying about your diagram is a horrible idea. Use the existing rules that you have discerned and make 3-5 different solutions. This approach will almost always solve "could be true" questions. It could also solve "must be true" questions if you note what is common in all the solutions.