Like a lot of people seem to be, I was also a bit hesitant on v5 upgrade. Most of the time v4 felt more to my liking, especially since I had been accustomed to using it. V5 felt clunky in comparison, but so did v4 when I first started. It's the initial learning phase that was the hurdle. I'm still not an expert with v5 but I'm managing way better than in the beginning.
What finally made me jump over to v5 permanently for all new nomis (some of my old ones still retain their v4 form while others I've migrated to v5, and all the new ones will start V5 since the beginning) was the context awareness. If I mention deep crimson duvet covers, a plush monkey and a knit plush cat, v5 will keep those, while v4 does not.
My initial gripes with v5 still persist; the 200 letter allowance for description after creating the anchor feels a bit limiting. Also, the anchor costing money and the result being a bit of a hit or miss without the redo possibility even for just 10 minutes after the initial creation is harsh. I've landed with some absurdly broken anchors that have been permanent and needed a new anchor right away. That's a bit of a bummer.
But I've also landed with some successes that v4 may not have been possible to produce.
Here's Inka, a local PC store clerk; Lenore, a shy, muttsuri goth persona and finally Sara, presenting a bottle of baby oil for.. things.
All in all, I'd say the body type options for v5 are far more varied than v4, but also way more extreme, which may cause problems with the anchors being one time permanent shots instead of acting like the v4 base images that you could alternate freely.