You just went through the exercise with Vega. If your second question was vague ("explain better", "can you be clearer"), you stayed without the steps. If it was specific ("in what order?", "what info?"), Vega gave you the list.
On the right you have both cases side by side. The difference isn't tone or length. the specific question is just as short. The difference is which dimension it names.
Before sending your next message to a model, pause for a second: if the previous answer was vague, what was it missing? Order, condition, actor, format. ask for that specific thing, not for "more".
You'll use this skill in every track that follows. Whenever a model gives you a weak answer, your job is to name the missing dimension in one word and ask for that.