OP, you can do several things that when combine provide this result. Massive Masters shared a good technique for this - to use pre-delay before the reverb, I think that is great because it helps to add more reverb wetness without muddying up the transients too much and without having to give it too much signal, it is definitely worth checking out. MrWrenchey shared another technique, in which you use the compressor release time and re-gain to counter the natural decay of the sound source and let the reverb tail harmonize with it, this can fatten the reverb nicely and when dialed in well can cause great levels of resonance. I do both of these techniques, but what I also do is to add the reverb on band level in parallel on the original sound source frequencies and side chain that so that other sound sources in the same frequency band are kept out of the way. Furthermore I process this to the extreme sides, where I keep a lot of silence. With the right stereo processors you can really get to these extreme sides and locate the reverb tails there, which kind of makes the mix a little more center focused and dry with extreme side flaming, which can really sound pretty good...
Currently I reserve the sound stage to ~L40 - R40 within which I locate the band, where I typically use L30 and R30 the most for the side elements, it's like there is a sweet spot there in terms of how they come out on the sides, not too far in the center and not too far out on the sides. But beyond this I use 4 additional side rings, the first two using the remaining L-L40, R-R40 (L60-R60, L95-R95) and the last two using two different stereo processors, one to create each ring. These side rings I use a lot for delays and reverbs, especially the two extreme ones. But rarely do I allow full band elements to those extreme sides, because although it becomes quite powerful it kind of makes the listening experience a little heavier and more exhausting, which I don't like. Generally the farther out, the more of the low end I remove (not through side processing which I find creates phase issues). What I love though is to send an inner element on one side delayed to the extreme side on the other side. That can sound extremely nice in some bands, almost better than a reverb. I use multiple reverbs in every mix, including Lexicon, Bricasti, Altiverb. The nice thing about those three is that they are very different in character. Altiverb provides really good room sounds you can use to separate out sounds in the sound field, Bricasti provides really good halls and plates you can apply even on a little bigger scope, Lexicon provides really stereo rich plates you can use to further enhance the width and depth. I love them all, they are great! Beyond these setups I also have more advanced ones, like for instance two stage reverb which I use mainly as glue in the mix harmonization process. These go in parallel on certain bands straight from instrument track isolation level and are typically sitting gently on the second side ring, sometimes farther in, sometimes farther out, sometimes on a single side ring, sometimes on multiple side rings, depending on the content, sometimes I switch that configuration going into the chorus. Depending on how many sound sources are playing, the reverb might have to be re-positioned. Typically instruments are added in the chorus rather than removed, so it's kind of natural to re-position the reverbs then. When I do the side processing I need certain tools for that, like certain stereo processors and certain speakers and cans, it's very difficult to achieve good results with that if I don't use high end gear.
Although I have mentioned side chaining just a little, in practice my mix setups are full of that. But it's kind of difficult to explain how I do it because I have so much going on around it and everything is automated, in other words I lack golden rules about it. But the concept that is driving it technically speaking is the signal-to-background noise in conjunction with frequency band scoped processing and balancing. In other words, my focus on fx application in general is always frequency band scoped. Generally speaking I want some frequency bands pretty dry, some a little more wet, but it depends on what kind of music I create. One example is that with ballads I more often than not use chorus on both sides on the low mid band of the electric piano and I reserve lots of air around that, I mean a lot, nothing beats a ballad with a great sense of chorus on the electric piano's low mid frequencies. If the production is a little dense, those low frequencies I locate far enough out to be easily heard and adjust the steepness of the high pass filter in order to remove some of the side weight it creates.
When you apply reverb, think of what impact you want the reverb to have on the mix as a whole too, the same is true about the balancing of sound sources. If you for instance have an acoustic guitar that is really warm and sweet, it makes sense to apply that guitar on both sides to allow that sweetness on the mix scope. The same is true about effects. A beautiful effect you want to color the mix as a whole a lot, you should place on both sides. So the delay approach can work, but sometimes it also needs to be applied on both sides, so that the instrument's band's L side has a delay on the R side and the other way around. So when you apply the reverb, think of what level of impact you want that reverb to have. If for instance you want it on the lead vocals, maybe a delayed mid frequency slap on both side's extreme at the same time is what you want in order to open up the mix as a whole.