Call backend functions on Plots.jl (PyPlot backend)

Is there a way to modify a Plots.jl plot by directly calling the backend’s functions on that plot? In my case, the backend would be PyPlot. I need to have a multi-column legend, which doesn’t seem to be supported.

For the pyplot backend you can access matplotlib figure objects from a plot p with p.o or axes objects with p[1].o and use matplotlib functions to modify them. An example can be found e.g. here: https://github.com/JuliaPlots/Plots.jl/issues/394#issuecomment-708047020