I think the confusion may be that usually when you have two bands RF engineers gravitate to 'diplexer' and there are a lot more questions to be answered, such as passband width, stop band attenuation, maximum acceptable insertion loss, transition band, etc.
With that said, there are several resources available on the web to get you started. You mentioned you need to fabricate one, but didn't mention your budget, so I'll approach it several ways.
The first is the old fashioned way, by using a Smith Chart, a decent intro is here:
pros: Its' free, relatively easy, if a bit tedious.
cons: The optimization techniques are limited in terms of bandwidth. Did I say tedious?
The second is a computer-assisted matching and there is a plethora of software packages (from free to Rolls Royce pricing).
pros: much easier than the manual technique, various data import options, various matching optimization algorithms and options (especially on the $$$ ones).
There are some free or low priced packages out there such as:
and more, I have no horse in this race BTW (as a disclaimer).
You can literally google "RF impedance matching software" and they pop out by the dozen, you may want to check reviews or send them an email to see if one would meet your specific tuning requirements.
cons:can be pricey, optimization controls can be limited for the cheaper programs, unlike the $$$ (Genesys eg).
But if you're not in the RF business where you're not trying to squeeze that last .02dB to beat the competition that may be enough.
There is a point of diminishing returns once the matching network becomes large, in that you're losing more signal than the matching network is allowing to pass (due to parasitic losses).
As a historical reference: