Citation Issues

I wanted to ask a couple of very basic questions about the high-level structure of genealogical citations. I need to make sure I "cover my bases" before discussing certain topics in more depth within FHISO (see below). I hope, too, that others benefit from the answers I'll receive  ;-)

1) Can you direct me to a succinct but accurate definition of a "layered citation"? I'm especially interested in whether they also apply to source-of-source issues as well as derivative editions.

When citing census pages for England & Wales -- whether on Ancestry or findmypast -- I tend to use 'database' to indicate the nature of source being consulted. I used to specify 'digital image' but was informed that this is incorrect.

I'm crafting a citation for a source that doesn't seem to neatly fit into any of EE's many categories nor into any hybrid combinations. The original record is a straightforward deed that would follow the basic formats in section 10.5. These records have been digitized and made available through computer terminals at the recorder's office but without Internet access. So it would seem that a layered citation is required, but both layers are supplied by the same agency. This is what I came up with as a citation:

Don't we all just love Ancestry.com and using their stuff in our research?  I have found a family tree there that I am referring to in a client report.  Trouble is, no matter many times I hint that I would like the submitter's name, it is not given.  Sooo...  I have the following citation that I constructed using the template for Documented family tree data in the "Citing Ancestry.com Databases and Images" Quicksheet:

I have my original Texas Marriage License that was returned after being recorded by the county clerk.  I am not sure how to cite this record.  Do I cite it as a privately held item or do I cite it as a loose papers government record following EE 9.5?  Or, another option is do I make two separtate citations, one as a privately held artifact and a second as a marriage record as it is recorded at the county office from the back of the certificate where this information is documented (EE 9.4).

Could this work as a citation?

I am a couple of weeks old in experience with Evidence Explained, so please bear with me on this. I’m sort of feeling like a ditz here.

As an example, I want to use the following source: https://familysearch.org/pal:/MM9.1.1/VC19-8GQ

In this one record, I have:

1.       The marriage date for Johann Heinrich Metzler and Anna Katharina Seibert

2.       The location of their marriage.

3.       The birth date for Johann Heinrich Metzler.

4.       The birth date for Anna Katharina Seibert.

What's the consensus regarding citation of a prior work by ones self? For instance, if you have previously made a particular case, or a reasoned argument, and you wish to make reference to it in a new work.

In scientific and mathematical works, there would be no difference, but I wanted to confirm this in the specific realm of genealogy and historical research because the subjectiveness of an authored work rates less highly than more objective sources, and for fear it sounds like the I-disease.

I’m in the process of cleaning up 30+ years of research. 

My current problem stems from not always documenting when I accessed records.  However, I have hundreds of media file images of these source indices, vital + census records, etc. 

When I restored my media files after a computer crash last year, it dated all the media files to the date of my restore, 28 April 2013. 

My question:  What do I put for the accessed date for the images no longer available or not available in the same format as the image I have.

Accessed date unknown

I have a delayed birth certificated for my grandfather issued in 1942 with his information.

In the early 1950's, his first cousin who had the same name, same first name of mother, same birth year and month and the day of month different by 15 days, applied for a delayed birth registration and my grandfathers certificate was changed to reflect the new data.  Yes, it is very confusing but it happened.