One disavantage of potassium soap is that it can take a long time to drop out compared to sodium, as a result often a mixture is used part NaOH and part KOH to get the benefits of both.
As your oil is still titrating quite high (6 or 4), I would be tempted to try with a little more NaOH/KOH and see if you can get that titration level down further, maybe only 5g of NaOH, or 3g KOH (if you have the samples left). The reason being that when you add alot of caustic (soda or potash) there will be a certain amout of neutralisation, but also an amount of direct saponification where the oil is turned directly to soap. The presence of the glycerol should help reduce this, but given the titration values you now quote it does appear that is happening.
If you remove the glyc, the oil will still contain some soap so trying to transesterify immediatey is likely to produce yet more soap and also give a very incomplete reaction. The problem with an incomplete reaction is that it can result in soap/methanol remaining in the oil/bio and maybe not dropping out at all. More likely to happen with KOH as that tends to take longer anyway.
Due to the high amount of caustic used wit the oil, it is quite likely to go dark, most should drop out with the glyc, but the presence of pattially reacted oil in the form of mono and di glycerides will tend to keep lots of things from dropping. As you get closer to a pass, you should find the glyc drops quicker and the bio gets lighter. The final bio is still likely to be darker that you are used to though.
PS.
I went upto Father's on Teeside as planned and only been there 20 mins before getting a phone that our 10yo had collapsed and an ambulance called - he is home and seems to be none the worse - but I obviously had a change of plan and returned early.
