If it's not memory pressure, then there may be ways to improve it, or may not. It really depends a lot on the data being queried, and on the complexity of the query itself.

But it may not be something that needs to be improved. First, identify if that's the actual performance bottleneck in the query. More likely, there are issues with the way the query is written, with indexing (missing indexes, indexes with poorly chosen leading edges, non-covering indexes, etc.), with index/table fragmentation, and so on.

99% or more of the time, worrying about the internal engine methodology for dealing with the data is much less productive than reviewing more obvious performance issues.