Since the water is doing well, have you considered an oil to water heat
exchanger. That will probably drive the water temps to 'problematic',
but then you have the option of implementing John's augmenter.

Yes; my earliest designs included oil to water heat exchanger because there
are some advantages to have them coupled.  But the KISS principal wins out.
I could find no oil/water exchangers that were designed for the high oil
flow rates of the mazda engine, so it could only work in a parallel
arrangement.  I did not design the coolant system to handle additional load
from the oil.

Al