Isn't the biggest usecase for B-trees in environments where we don't hold the data in memory (i.e. databases)? There the block size is optimized to balance the

Isn't the biggest usecase for B-trees in environments where we don't hold the data in memory (i.e. databases)? There the block size is optimized to balance the cost benefit of one more disk seek vs one more list traversal.