There’s growing evidence that the Anatolian branch - long thought to be the earliest split - may be better described as a sister branch. If so Proto Indo-Anatolian would be the family, with an Indo-European branch and an Anatolian branch.

I believe this has been a minority position among linguists but recent aDNA data suggests that sister branches is a good fit for the genetics, giving a little more weight to that model for the language.