Yes, helps with data skipping.

From docs: If you expect a column to be commonly used in query predicates and if that column has high cardinality (that is, a large number of distinct values), then use ZORDER BY.

https://docs.databricks.com/en/delta/data-skipping.html