1) as you said, its flux is predominantly in the IR
2) it could have been fainter than the sensitivity of the HST instruments but now seen because of the sensitivity of JWST
But given that it appears so bright in the JWST image similar to other nearby galaxies that do appear in the HST image, your bet on #1 seems reasonable.
Also there is another point: rather than a highly redshifted galaxy it could be a very dusty nearby galaxy (also appearing very red) but if I remember right, that would have a slightly different signature. Dusty galaxies often aren't entirely dusty and have "lanes/channels/streaks" of dust that are interspersed among normal stellar regions, so if it were that, you would be seeing some bright spots outside the infrared. But this one has the shape of a normal galaxy but red all over, suggesting something affects the whole galaxy -- i.e. redshift.
Do the recolored images have any relation to what the original view would have looked like, or is it just arbitrary "artistic license"?
They did not release or talk in much technical detail of how the images were assembled, which I'm sure will be done at some point.
I do not think the colors do correspond (at least not deliberately), for 2 reasons:
First is that the image of the Deep Field ("SMACS") contains galaxies at a range of redshifts. For example there may be galaxies quite near us (redshift z = 0 or close to 0) while others are more distant (the arced galaxies in the image being lensed that this image is famous for, at redshift z = 0.39), where redshift is the measure (1+z) of how much the wavelength light has been multiplied.
So regardless of what color mapping you chose, it would not be a perfect fit for all objects in the field of view. For the galaxies near us in the image (z=0), the wavelengths being converted to RGB don't correspond to what we would see in the optical.
Secondly, if it were remapped especially for the galaxy cluster of interest in this image, I don't think the colors are specifically tuned for that either.
In more detail:
Consider the optical color spectrum we see, ROYGBIV, or let me reverse it in order of increasing wavelength VIBGYOR -- and take the "RGB" 3 colors that might make up an image, or BGR to use that ordering -- this spans a wavelength range of say 400nm, 600nm, 800nm.
The imaging filters available on the NIRCam span 900nm to 4400nm (4.4 micron) and there are 29 of them [0]. Researchers choose which filters to use based on what they wish to study. And recall that the imaging sensor actually outputs grayscale only, it is the filters that give it a color view and individual images in each filter are assembled to create a color composite.
According to an example science program designed to take such images[1], the filters selected to be imaged might be 900nm, 1150nm, 1500nm.
If you applied the redshift of the galaxies (divide by 1.39 from the above info about the cluster of galaxies), the above sampled wavelengths in the image would still correspond to redder parts of the spectrum compared to what is visible if we were seeing the galaxies now: 647, 827, 1079nm.
So, no I don't think the color mapping was chosen to be accurate in a scientific sense of seeing what you would see if the galaxies were brought to the "original" view.
[0] https://jwst-docs.stsci.edu/jwst-near-infrared-camera/nircam...
[1] https://jwst-docs.stsci.edu/jwst-near-infrared-camera/nircam... ("select filters")
This kind of mimics the human eye, which is sensitive to those three approximate color frequencies, but it's interesting to note other species besides humans (and, apparently, even some humans)[1] have vision that work with more "filters", or on different spectra (such as honeybees).
I've always found it kind of amazing that so many satellite imaging devices work on far greater spectral ranges with far more color filters, being able to discern far more information than we could with the naked eye (but in essentially the same way).
However, 2 factors:
1) there is an intrinsic limit I believe to how much more resolution you can recover (maybe a factor of approx. 2x?), for a lot more exposure time needed However, also at these faint levels of brightness you're also competing against intrinsic photon and sensor noise)
2) practically, given the value of the telescope's time and not much more to be gained (science-wise) from achieving this next order of spatial resolution, they want to spend the time on other new targets instead of sitting on the same patch for much more time.
(you can even try this at home: https://petapixel.com/2015/02/21/a-practical-guide-to-creati... )