The answer is not straightforward: although there are parameters for D3 compatible with MN06-L, the literature does not suggest that the introduction of dispersives will necessarily lead to an improvement in the description of the system. This is because Minnesota functionals are parametrized on a database that already contains non-covalent interactions (Article How Accurate Are the Minnesota Density Functionals for Nonco...
). Probably the best procedure could be to search the literature for benchmarks on systems similar to those targeted.