Why do you think XUL and Thunderbird are good investments vs. things like Servo or the context graph initiative (which the Pocket team is going to accelerate according to the blog post)?
Do you think a non-standard framework for making fat native apps and a mail client built on this framework are going to make Mozilla more relevant?
I think it's a failure of imagination to assume that this is going to be the extent of what Pocket will be able to contribute... they have a lot of users and popular mobile apps for instance which is useful on its own.
As the blog post points out, the hope is that Pocket will accelerate the context graph initiative.
So, I disagree that it's about "bloat" or customization. In any case, the Pocket add-on is like ~700k and it doesn't do anything if you take it off the toolbar...