Hi,
The solution to choose will really depend on what are the business priorities ... If you are really looking for a cost effective solution I would go with option 1 which is a very common VVR setup to have 2 nodes in prod in VCS while having 1 node in DR. With this setup you can implement GCO & make your data group as global service groups which will have capabilty to failover between sites in any disaster event.
More consistent though not cost effective solution would be to use Storage based replication like SRDF or Hitachi TruCopy. This will ensure more consistency though cost will be higher as comparing to VVR.
With the purpose of disaster recovery in mind,in the same cost what you will put for SRDF or Trucopy, you can build an third DR site with VVR with lesser cost & keep two copies of data. You can achieve this by creating another site with one node (whether GCO capable or not) with primary directly replicating to two different sites OR creating a bunker to replicate to two different sites.
Details on Bunker replication can be found here
https://sort.symantec.com/public/documents/sfha/6.0/solaris/productguides/html/sfhas_replication_admin/ch15.htm
G