Are there use/edge-cases for implementing the iterator design-pattern yourself, when iteration-supporting-interfaces are available?

The iterator design-pattern is employed to sequentially access elements of a collection without exposing its underlying representation, as is explained here and here. Since C# has interfaces that support iteration, are there still use/edge-cases where it would make sense to employ the iterator design-pattern yourself, such as is done here, even though available interfaces already implement the iterator design-pattern?