The common understanding for Accuracy and Precision in English language is given by this figure.

Inspired by this question I have a follow up question relating Accuracy and Precision in Mathematica and Wolfram language.

How do we understand the relationship between Accuracy and Precision in the following example?

Accuracy[
 SetPrecision[
  SetAccuracy[
   12.3
   , 20], 15]
 ]

13.9101

Precision[
 SetAccuracy[
  SetPrecision[
   12.3
   , 20], 15]
 ] 

16.0899 

Where is this 13.9101 and 16.0899 coming from, exactly.

Given an operation, such as Subtract, Plus or Times.

How do we predict the Accuracy and Precision of the outcome?

Precision[
 Times[
  SetPrecision[10, 3] ,
  SetPrecision[1, 7]
  ]]

2.99996

Precision[
 Plus[
  SetPrecision[10, 3] ,
  SetPrecision[1, 7]
  ]]

3.04139

Accuracy[
 Plus[
  SetAccuracy[10, 3] ,
  SetAccuracy[1, 7]
  ]]

2.99996

Accuracy[
 Times[
  SetAccuracy[10, 3] ,
  SetAccuracy[1, 7]
  ]]

2.99957