One could lump epigenetics in with "environment", but there are three reasons for not doing so.

(1) Many epigenetic changes can be detected in the DNA, just as methylation patterns not sequence variation. So it's worth splitting this out as a third category.

(2) Epigenetic changes can occur over longer timescales than one normally considers environment to be important (e.g. inherited from parents), so it's awkward to try to control for it.

(3) Epigenetic changes are read out as expression pattern- and level- differences, which in many cases are equivalent to altering a protein's activity or baseline expression level, which is what allelic varation often is. So it is the same in-kind as the genetic changes; it's just implemented through a different mechanism.

Nonetheless, I agree that for simplicity one could leave it out, especially since the results are a lot more intriguing than well-established at this point as to what difference it makes.