You may have trouble seating all three of those bullets "optimally" with a single freebore length chamber. I generally try to keep the seated bullet boattail/bearing surface junction in a region above the case neck/shoulder junction, but not more than halfway out the neck, to where I will have distance between the seated bullet ogive and the lands of something like .025" to .030" off, to as far as about .010" into the lands. A .223 Rem case neck is very close to .2" long (0.2027" by SAAMI specs), which means there is ~0.100" seating depth "window" between the case neck/shoulder junction and the neck midpoint. In other words, you have about 0.100" to play with if you want to keep the boattail/bearing surface junction within this region.
That doesn't mean you can't seat a bullet a little deeper with the boattail/bearing surface junction below the case neck/shoulder junction, or with it a little farther out than the neck midpoint. However, both of those scenarios may bring different issues you'll have to deal with if you choose to seat bullets that way. For example, the farther down in the case you seat the bullet, the less usable case volume you have, which might lead to insufficient powder capacity and/or pressure issues. Bullets seated very far out in the neck may require greater neck tension to keep them seated with sufficient force. Again, these may not be deal breakers, but something you have to deal with nonetheless. How severe the issues might be would largely depend on how far you had to move the bullet past either boundary.
PTG optimized a reamer for the 90 VLDs years ago that cuts 0.169" freebore. At the time, most people shooting the 90s were seating them into the lands anywhere from touching to maybe .010" in. In my hands, they shot very well from about .004" to .007" into the lands in a 0.169" freebore chamber, but it was a very narrow optimal window. By experimentation, I found they also shot very well around .020" off the lands, but it was a much wider window. By going with that seating depth window, I gave up almost .030" of the usable freebore as compared to those shooting them at .010" into the lands. The boattail/bearing surface junction of 90s seated at .020" off in the 0.169" fb chamber is just barely above the neck/shoulder. That means you have to run higher pressure to achieve a given velocity than if the bullet were seated further out in the neck. In a 30" barrel with H4895, the sweet spot for the 90s to tune in is in the 2850 fps range, which is very hard on brass. That extra pressure makes the problem even worse. IMO, the 0.169" fb is not sufficient for the 90 VLDs if you are seating them off the lands. You can certainly get around that by using a different powder such as Varget, which gives slightly less velocity, but at lower pressure. However, you're giving up a small amount of performance and Varget has larger kernels than does H4895, which generally makes charge weight variance a little bit larger. You didn't mention the specific type of shooting you would be using this setup for. In F-TR competition, many are using .308 Win with 200 gr high BC bullets, which means the .223 Rem with 90s will already be at a disadvantage in terms of wind deflection. In addition, the smaller case volume of the .223 Rem means you'll usually get almost double the velocity variance as compared to a .308 case simply because any charge weight variance will have almost twice the effect as it does in the larger case. The substantial reduction in recoil with the .223 Rem is certainly worth something, but the point is that you can't afford to be giving up too much in light of the disadvantage the .223 Rem has from the outset in F-TR.
To [finally] get to your questions, the lengths from base to ogive for the 3 bullets you listed are 0.541" (SMK 80), 0.589" (Berger 80.5), and 0.622" (Berger 90 VLD) [Bryan Litz, Ballistic Performance of Rifle Bullets, 2nd Ed., 2015]. To optimally seat both the longest (90 VLD) and shortest (80 SMK), you're already using up most of the ~0.100 neck window for optimal seating, simply due the differences in BTO lengths, so the relative seating depths at which each bullet tuned in could also have an impact.
The obvious choice would be to have the rifle throated for the 90s, rather than for the middle or shortest bullets, to prevent having to seat any of them too far down in the case. I have seated the 80.5s in my rifle chambered with 0.169" freebore at ~.020" off the lands. It's doable, but they're seated pretty far out in the neck; I wouldn't want them seated much farther out. Because SMK 80s are even shorter, I'm not sure if they're are quite long enough to have sufficient shank gripped by the case neck and still maintain the necessary distance relationship to the lands. That would largely depend on where they actually tuned in with regard to seating depth. If it turned out their optimal seating depth was only around from .010" off to somewhere into the lands, there wouldn't be a whole lot of shank left in the neck. A better choice would be Sierra's 90 gr bullet, which has a longer BTO.
Finally, a 7-twist is sufficient to stabilize the 90 VLD, which is the longest bullet of the three you listed. You may [in theory] be giving up a few percent of the maximum intrinsic BC, but it has been observed that running the 90s in the 2800+ fps range with faster twist rates (i.e.< 7.0) can lead to jacket failure, something you really never want to experience during a match. A 7-twist is plenty, and apparently not quite as hard on the jackets at higher velocities.
