Aside from one type of disk bottlenecking the other, are there any other problems with mixing SSD models in RAID?
My problem is, I need to upgrade the storage in a server with 4x Samsung 845DC EVO 960GB in RAID10. These drives are not available anymore, so my options are to either use some newer comparable SSD's or to replace the array altogether.
The single biggest thing that crosses my mind isn't SSD-specific: that the biggest danger with RAID is that all the devices in any given RAID are often purchased from the same manufacturer, at the same time, and therefore tend to get to the far end of the bathtub curve and start dying at about the same time. In that sense, buying from different vendors is not only not a bad idea, but best practice.
You don't say whether you're doing hardware or software RAID. If it's hardware, you have the issue of whether the new models are supported by the controller, both from a hardware support contract standpoint and an "it's too new for me to talk to / my programmer told me not to talk to you" standpoint. Either of those would be a reason not to do it.
There is also the issue of capacity: if you're adding devices that are smaller than your existing ones, even if by only a few sectors, this will not go well. Check the absolute raw capacity to ensure it's greater than or equal to the devices you're already using.
But assuming you can get past those issues, I think it's generally a good idea to do what you're planning.
In fact as SSD tend to die from heavy use, and as an array of SSDs in a RAID are submitted to exactly the same write pattern, I try to mix different models while keeping close; for instance, I mix Intel 530 and 535 SSDs. They're different, but close enough for it not to be too much a problem because of too different behaviours.
If I can't mix SSDs from different models, I write lots of data to some of them in various amounts, to get them "primed" at different levels (you have to write at least a couple of terabytes on each SSD).
