Peter
Can we separate your issue into what I see as two issues:
- The Genealogical Issue of complex but uncertain families
- The recording of such families within FH
1. Uncertain Families
You have
- a marriage under one surname,
- the births of some children with the same surname (or slight variation) in the 2 decades following
- a (possible) death of someone who might be the wife of the first marriage in the middle of that period
- a possible second wife with a very similar name to the first wife
- all events happening in the same English parish
- all events happening well before census records
There is a temptation to say that there is one father, who married twice to women of similar names and that both marriages had children. In the baptism of one of those children the surname was spelt slightly differently.
You need to decide whether you should succumb to that temptation or whether you want to look for further information (or even the lack of further information - which could be significant). Then you need to decide if your conclusions are certain enough to be recorded (if necessary with notes about uncertainty) in FH. You may never achieve full certainty.
You need to resolve the "was there a second wife" question. Do you have a burial record for the first wife and a marriage record for the second? If this is only based on baptism records changing from "dau of John and Susan Ablit" to "dau of John and Susanna Ablit", this could just be the couple wanting to sound grander, or a new cleric not abbreviating the mother's name.
You might look for a will of the father or the surviving wife (there may also be one for any first wife). In one of the wills I found, the testator had conveniently referred to "children of my surviving wife Maria" and "children of my first wife Sarah".
You might look for burial records or memorial inscriptions. If the father is buried in the same plot as both his wives ...
If the first wife is buried in her family plot ... Again gravestones or memorials which refer to multiple surnames often help.
If there are two wives, they were probably both baptised. So check the records for the parish and say at least the neighbouring ones for possible baptisms - can you "prove the existence" of both these brides?. Do those families then have baptisms of siblings of these wives and are their Christian names then reflected in the family you are investigating (not "proof" but kind of reassuring).
In the parish records are there lots of Ablits - in which case you have to consider multiple families? Do the baptism records give an abode or calling for the father? If these are constant that is an indication that you may be dealing with the same couple. (If they are not, it is not as easy to conclude that they are different couples - they may just be "going up in the world" (or down)). If there are lots of Ablit(t)s, you may want to try and do a reconciliation of all parish register entries for Ablit(t)s in that parish and neighbouring ones for the period plus/minus 10-20 years. That could be a massive and inconclusive task - but it might give you a warmer feeling that you have one family. Remember that the situation may be complicated by step children (such as children of the second wife assuming their step father's surname in the marriage register).
Sometimes the Bishop's transcripts are worth examining as they might contain additional detail. It is not unknown for the "transcriber" to add his own opinion of some of the goings on - particularly in relation to baptisms and parentage - to the transcript. Likewise if your original source is actually a Bishops Transcript (are they all in a suspiciously similar hand, constant state of inkiness, are the entries manually numbered - indications of a BT), go back to the actaul parish records to see if there is additional information not carried through to the BT. (I appreciate from NZ that might be difficult!)
2. Recording in FH
In the father's property box, you can add a second wife - which creates two tabs with the wives names on them. You then need to select the right wife before adding the children. If you have the death date of the first wife or the marriage date of the second, that would help you assign the children to the correct mother (if the baptism records don't specify). Remember that the second wife may have been pregnant on marriage.
In respect of the surname variation, it is possible to add name variations. In the property box to the right of the Name there is a "more ...", clicking on this takes you to a dialogue that allows you to add additional names.
If you are using FH as a pure repository of data, you can add as many notes as you wish to document the uncertainty and you reasoning for the various relationships. You may want to put these in a single shared note attached to all the people involved. Make sure the first line of the Note makes clear what it is about so that it is readily identifiable.
If you are using FH as "an engine" to drive automated reports, CD, books, websites etc. You have to decide whether the information is sufficiently certain to even enter in FH or whether you can include sufficient indicators of uncertainty that will make it through to the output and convey your uncertainty.