The target is the problem you choose
Start with the drug target. Picking a target is choosing *where in the biology you intend to intervene*. Suppose a disease is driven by an overactive protein. That protein becomes your target. Notice that choosing the target says nothing yet about what your drug will be — only about the problem you've decided to solve. Whether a target can even be hit by a drug at all is its druggability, and some very important targets are stubbornly hard to drug.
The modality is the kind of tool
Once you know your target, you choose a modality — the *broad class of therapeutic* you'll build. Think of it as deciding whether to use a hammer, a screwdriver, or a wrench, before you pick the exact one. The two most familiar modalities are the small molecule (a tiny, often oral compound) and the biologic (a large molecule made by living cells, such as a monoclonal antibody given by injection). Each modality has its own strengths: small molecules can get inside cells and be taken as pills; biologics can be exquisitely specific and hit targets on the cell surface.
The drug is the specific tool
Finally the drug itself: a single, exact molecule of the chosen modality, aimed at the chosen target, working by a defined mechanism of action. The mechanism also tells you *which direction* you push: a drug that switches the target on is an agonist; one that shuts it off is an antagonist. Same target, opposite intentions — and that choice is part of designing the drug, not the target.
So the sentence grows: *for an indication, you pick a target, choose a modality, and design a drug that engages that target by a mechanism.* Beginners who keep these layers separate avoid 90% of the confusion in everything that follows.