I will make a few comments based on my I understanding and personal observations.
Cryo is a treatment in cryogenic liquids. Liquid Nitrogen, at -321°F, is the most common for knifemakers to use. It will reduce RA by making it convert to martensite, as well as convert some carbides into eta carbides.
Steels with hypereutectoid carbon and sufficient alloying are suitable for cryo. The changes in cryo are time dependent, and require several hours at the -300°F temperature to occur. Six hours to overnight is the normal cryo treatment.
Sub-zero treatment is submersion in a liquid that is chilled to -95°F or slightly lower. This is normally a bath of crushed dry ice and denatured alcohol. Some use dry ice and acetone, but the volatility of acetone is much greater than alcohol, thus making alcohol safer and cheaper.
Sub-zero brings the RA down to the Mf and converts it to martensite. It does not affect the carbides. The steels that do best with this treatment are ones with carbon content closer to the eutectoid. AEB-L is a good example of this type steel. More complex steels that would normally benefit from cryo will still reach full hardening and minimal RA with a sub-zero treatment.
If liquid nitrogen is not available, all stainless and high alloy steels should at least have a sub-zero treatment in a dry ice slurry. When the Mf is reached, the conversion is done. This happens as soon as the steel is at -95°F. Longer soaks will gain nothing.
Freezer/Freezing treatment is placing a quenched steel object in a freezer at or slightly below 0°F. This is usually referred to as a cold treatment. It is best if the freezer is -10°F. This treatment is a gray area that gets hyped by some and poo-pooed by others. The Mf on simple knife steels is at or above room temperature. Nothing more happens by cooling the blade to below the Mf. As more alloying and carbon are added to steel, the structures get to where the Mf starts to drop. In stainless steels, the Mf is around -100°F. In a few alloyed carbon steels, it can drop to between room temperature and 0°F. Steels like 52100 and some other alloy bearing carbon steels will benefit slightly from assuring that the Mf has been reached. Placing in a freezer for an hour or two to bring the blade down to Mf is OK, but longer times ( often given as "overnight") will do nothing more. The difference between a blade quenched down to room temperature and one done to -10°F is probably nearly indistinguishable. Looking at the industrial charts shows that only a few knife steels have an Mf between 0 and 70°F, so doing a freezer treatment is of little use for most bladesmiths. Looking through the ASM guides shows that none of the standard alloyed steels ( A2, D2, L6, O1, etc.) call for anything but a room temperature quench. In short, it won't hurt, but don't expect any significant gain.
Of more importance is the timing of events in HT. Quench should be immediately followed by the next treatment, be it cryo, sub-zero, freezer, or tempering. More accurately, the treatment is the lower end of the quench, and the drop should be as continuous as possible. Delays in the drop, and in the tempering stabilize RA and make the next steps less effective. The first temper should immediately follow the quench and treatments. You may loose more in a delay in tempering than you will gain from any cooling treatment.
Observation in objective tests (cutting things and other use tests) is not very accurate for small differences. The objectivity is usually skewed by the mere knowledge of the treatment being tested. A test of this principle was done by a university in the 1960's where two identical cups of distilled water were offered to people to taste test. One was called plain water and the other was called "improved taste" water. Nearly every taster said the improved water was much better. In a truly blind test, you would not know which is which, and the results would be more accurate. Add to that, that performance tests are hard to keep identical, and there is even more room for error in observation.
I will compare this whole discussion to the current super moon. The word super implies a huge gain. The full moon is at perigee (at its closest), and thus appears slightly brighter and larger looking on a clear night. The actual difference is about 4% in brightness and 2% in size from last months full moon ( the 14% size and 30% brightness often posted are not an apples to oranges comparison. They are the difference between apogee and perigee, which happen about 7 months apart.). The difference between the moon at rising and when directly overhear on any night is almost the same difference as the super moon is from month to month ( the moon is about 3000 miles closer at meridian than at moon-rise/set).
The difference in the moon's appearance is much more than the gain on a knife blade from a cold treatment, which is often less than one percent. Without being told, the average viewer would never see the difference in the moon tonight as compared to December, and the average knife user would not be able to tell the difference between a 52100 knife quenched to room temp vs one done to -100F. But, if you asked them if the super-moon looked bigger, they would almost always say, "Yes, it looks much bigger". The same for the knife. Ask if the cold treatment knife seems sharper or longer cutting, and the tester will almost always say, "Yes, it is many time better".
