I think a good first step would be to familiarize yourself with the techniques you are using and the quantities your are measuring. It seems that the difference between absorbance and absorption is an important point here: https://en.wikipedia.org/wiki/Absorbance and https://en.wikipedia.org/wiki/Absorption_%28electromagnetic_radiation%29

plotting everything in one diagram requires quite a bit of data processing when you want to compare transmission UV/vis and ATR-FTIR spectra as the effective path length in ATR is not trivial (see e.g. Anal. Chem. 80, 3045 (2008)).

moreover, it is a bit weird to  talk about the absorption spectrum starting from 0 µm. the common UV/vis can go down to about 180 or 190 nm, but beyond this you need an x-ray source...

First of all, you usually don't measure absorbance! What you measure is either transmittance and convert the transmittance into absorbance (quite often in an incorrect manner, cf. link below) or the reflectance. What you actually have to do is to derive the complex index of refraction function from your transmittance / reflectance measurement the imaginary part of which is the absorption index which can then be converted into absorbance. For the reflectance measurements (ATR or external reflection) you have to use the Fresnel equations (e.g. by applying the Kramers-Kronig relations) and for the transmittance the Abeles formalism except if your sample is a solution. Then you can use A = -log (T/T0) where T0 is the transmittance of the pure solution. I hope this clarifies things...

Article Employing Theories Far beyond Their Limits-The Case of the (...

The answers above can be summed up simply that the absorbance value is related to pathlength. With ATR this varies with the relative refractive indices of theatre crystal and sample and every wavelength. It is not an absolute measurement.

You also asked about 0-6A. In 1994 members of the then ICI carried out a round robin  test with two film samples, on various FTIR spectrophotometers, in transmission, primarily to ascertain wavelength reproducibility but also absorbance consistency. From 0-1.5 A the results were good, 1.5 - 2A reasonable, above 2A very inconsistent. At the time one instrument manufacturer was claiming linearity 0-6A. However this was achieved by employing an automatic gain change.

Yes, it is also the pathlength, but it is important to realize that in a transmission experiment the pathlength is a function of the complex index of refraction of the sample due to multiple reflections at the interfaces. Therefore the pathlength should never be confused with the thickness of the sample...