The next thing should not be email at all. It should be something fundamentally different, not another iteration of email.
Seeing some downvote criticism on this comment, let me clarify.
I understand the value in working on an established set of federated protocols. Everyone has an email address and this is a very convenient way to start. But it was invented decades ago and has had many kludges bolted on to it to deal with problems or to very slightly improve the user experience. Who's instead going to be more bold and go for a greenfield project to do asynchronous personal communications?