On the choice of a third (and fourth) generalizing parameter for equations of state