Yes.

Successful research requires awareness of what is already known, in order to see gaps in our knowledge, or to recognise that something is new when it appears in front of our eyes, even if we are not looking for it (a.k.a. serendipity ).

Serendipity cannot be predicted but we can prepare ourselves for the possibility, even as we pursue research in which the outcome is predictable.

If 'needs' refers to social needs rather than abstract academic needs, then such needs of course do need to be known for applied research.... but I believe that those involved in applied research can also keep their eyes open for unexpected observations that might lead to a new basic understanding of something.

For example, if underground water pipes in a new surburb keep getting blocked by roots from certain trees, this observation could lead to a breakthough in future drain design generally, or in tree ecology generally, even if the aim of a study was to make a plan for maintenance of existing drains in the suburb.

Doctors, like engineers, can be doing basic research even as they do applied research on known needs.

Of course, what I am suggesting here for applied research also needs wide knowlege of a subject, and around a subject.

Which us brings us back to the need for open access publishing whenever possible!