There's no significant change in information. The change from fin to limb actually includes informational/gene loss (associated with dermal bone, for example). As counterintuitive as it may seem, gene loss is often a creative force and patterns of gene loss are widespread in the animal kingdom. But laying that aside the take home lesson here and one of the surprising discoveries over the past few decades is how contrary to expectation major phenotypic changes are often not associated with major genetic changes but tinkering with what already exists. We see this particularly with the developmental regulatory homeobox genes common to all animals. All animals essentially have the same developmental genetic 'toolkit' since before the Cambrian period.
(Regarding 'combinations' I'm not clear on what you're asking, but I would add that combinatorial genetics is also an important factor, where again it's less about mass amounts of new genes and genetic information. Great differences in phenotypes result not only by how the same genes are regulated but how they interact with other genes in additive and combinatorial genetics. This is how, for example, the same gene in one context will result in different phenotypes in another context just depending on what other genes it's associated with)
No significant change in information; one produces a fin, another an arm with a hand, okay.
I think he is asking the amount of limb combinations.