If standalone email clients really are dead for the average user, that to me would suggest Thunderbird should focus on power-users and/or corporate deployments,

If standalone email clients really are dead for the average user, that to me would suggest Thunderbird should focus on power-users and/or corporate deployments, not that it should shift its focus to being part of yet another "service". It's perfectly fine to make a product with an audience that isn't everyone on Earth.