You're right: predicting is the key goal, not "explaining". A similarly unhelpful word (which comes up a lot in proposals) is "understand". Often, "to understand X" often means little more than "to not be surprised by X". At one point, people explained and understood in terms of mystical thinking that we now regard as quaint and wrong-headed. A more robust approach is to predict, and then to check those predictions against new, or at least independent, observations.