I've also argued for designing targets around capabilities, rather than the other way around, and have argued that the level of domestic CH4 cuts proposed by Ardern-Shaw were poorly chosen - plucked straight from a table in SR1.5 that cautioned against using the table that way.

This is why we have been arguing that people should Indicate separate contributions of long-lived and short-lived greenhouse gases in emission targets... this completely avoids the conflation that Hannah is troubled by.
Ultimately the problem isn't GWP*, it's GWP100.

www.nature.com/articles/s41...

GWP* also more accurately reflects the warming from a pulse emission over time than do GWP20 or GWP100.

But this assumes its conclusion. Of course most the overshoot is from CH4 if you already assume massive CO2 reductions because you've told your IAM to do that. But wishing don't make it so.
Until we see real CO2 reductions, the Pierrehumbert 2014 argument seems pretty reasonable.