I find myself adding layers in RGB mode and then wanting to add layers in LAB mode. However, doing so requires me to lose my RGB layers due to required flattening. And once that’s done I’ve lost all records of the RGB work for later adjustments. How is everyone solving this issue?
I’m trying to get all my development workflow into one file. Is this realistic? Otherwise I can’t keep track of which file is related to which and I’m forced to start all over or partly over.
Igor, I am not extremely familiar with LAB processing but from what I have tried, if you use Smart Object you might be able to work around it.
A dialog box will prompt you to rasterize because it will affect the rendering of the smart object but you can select “Don’t rasterize” and proceed. The curve layer that I added after that shows that it is LAB. And if I double click the smart object, it will open another file that has all the RGB layers in tact.
I am not sure though the side effect that you can get by not rasterizing the smart object. Maybe somebody more experienced with this can help.
On a separate note, Affinity Photo can add layers in different color mode in one single document. You might want to give it a try.
You can also achieve this in the TKPanel if you have that. Under the TK button select RGB 
LAB which will do what Adhika described, create a smart object and will leave your layers intact in each.
@David_Kingham Just curious, if you go back from LAB to RGB, will it create another smart object that encapsulates everything that we have done (RGB work + LAB work)?
Just curious, all, but what are the advantages and/or disadvantages of switching back and forth between RGB and LAB?
–P
RGB gives you a greater choice of adjustments while LAB gives better colors renditions. I typically do RGB and then end it with LAB. This is based on recommendations from a book by Marc Muench and it seems to work for me.
Also, once you switch to LAB you can forget using ACR. You have to flatten your layers and convert to RGB first. As a result I try to use ACR as little as possible. However, I really like it Texture and Clarity adjustments.
When you use the TKPanel action it creates a new document that is in LAB and flattened, you then work on it in LAB and can even leave the layers intact, then when you go back to RGB it takes that new document and converts it to a smart object and brings that back into your original RGB document as a smart object layer, so it’s non-destructive. Hopefully that makes sense and maybe @Tony_Kuyper could interject.
In reply to @David_Kingham, when you use/click the RGB <-> LAB action the first time, it duplicates the image, flattens it, and converts it to Lab. When you use/click the RGB <-> LAB action a second time (while the Lab image is the active document), the action saves all your adjustments to the RGB smart object. So everything from Lab is intact within the smart object in the RGB image. You can double-click the smart object thumbnail in the RGB image and reopen the Lab image with all the layers and adjustment you made intact. And if you do reopen the smart object by clicking the thumbnail, you can again use the RGB <-> LAB action to save it back to the RGB smart object. Hope that makes sense.