5) What variables are or should be considered additionally to the central EVs and DVs?
These are variables that you might need to control, in the sense of 'exclude the effects of' (which I call CVs!). They may well not be mentioned in the research question/hypothesis, but are nevertheless crucial. They are things that may otherwise interfere with the results and make it hard to interpret what you discover about the central variables in the design.
You can 'control' or eliminate such variables in various ways. One is by making them constant. E.g. you choose only people in their twenties for a study comparing men and women, thus eliminating the age variation factor; for an experiment where people read two types of text (narrative and argumentative) you make all the texts at the same level of vocabulary difficulty. Another way is to randomise the variable (or, more often, claim it is as good as random, even though you have not strictly randomised it…): to eliminate age you pick men and women randomly of all ages, so hopefully you will not get a lot more older people in one group than in the other. We have already seen also the 'stratified sampling' solution to this sort of problem, where you would pick equal numbers of people of different age groups in each gender, and the use of the 'matched subjects design' which also eliminates this, if age is chosen as one of the variables to be used for matching